[Pw_forum] SCF convergency

[Eyvaz Isaev]

Hi,

>>Why you think about a local minima?
> 
> Because in my previous unconvergent run total energy
> had ever reached -5865 Ry, > much lower than that of
>current convergent run. My  co-worker used annealing
> method in a MD run with CPMD to calculate the same
> system and got a total  engergy  of about -5820 Ry.
>
I am not sure that total energy from 2 different
programs could be compared directly. It depends on
pseudopotentials, cutoff energies, k-mesh, etc. 
 
> >
> >For test purpose, I assume, you can check lower
> >conv_thr, too.
> 
> I'll try that:-)
OK.

> 
> >
> > > Probably, I need to start the calculation from
> > > another point to avoid
> > > the local minimum. How can we do this in PWScf?
> > >
> >Not so clear what what do you want. Please clarify.
> 
> I was comparing scf calculation with an optimization
> problem of  multivariable function.
> Mathematically, when we got problem to reach a
> global minima, usually we  need to
> change a little our initial guess to start the
> search from a different point  in the relevant
> parameter space and see if it makes any difference.
> 
Well, you can change your experimental lattice
parameter (increase and decrease it by, say, 5%), and
then calculate and compare total energies.

Bests,
Eyvaz.


		
__________________________________ 
Yahoo! FareChase: Search multiple travel sites in one click.
http://farechase.yahoo.com